Model techniques map
Techniquespost-trainingreinforcement learning algorithm

implementation detail · filed under post-training

Zero-Reward Correction for Leaked or External Answers

Resetting a trajectory's effective reward to zero when evidence confirms reliance on an external or leaked answer.

source
1
models
2
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

When evidence confirms dependence on an external or leaked answer, we reset the trajectory’s effective reward to zero and treat it as a failure before recomputing group statistics.

usedpost trainingin MiMo-V2.6-Pro and MiMo-V2.6-FlashXiaomi

Filed alongside

Other methods under post-training :: reinforcement learning algorithm.